• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

EL84 Amp - Baby Huey

My BH monoblocks are each powered by Hammond 270EX transformers 275-0-275 @ 125ma each; the Hammond website now rates the 270EX at 144ma or 118VA.

The EL84's need about 38ma each x 4, plus the phase splitter current, etc. so a stereo amp needs 175-200ma or so....more than 100VA
 
Last edited:
I visited my old dear at the weekend and had a look at her BH.

Couple of points, the bias for the input stage seems to drift every time I visit. Reading the plates of the input valves it always drifts a few volts. What might be the reason for this? Or is that quite normal?

Also, a more generic question. The volume pot becomes noisy from 50%-99% turned, becoming more and more noisy. Then at its fully on position it is dead silent. Is this poor layout or wiring? Would a stepped attenuator be the answer?

The sound still amazes me, can't believe I gave it to her! :headbash:

Cheers

Charlie
 
Couple of points, the bias for the input stage seems to drift every time I visit. Reading the plates of the input valves it always drifts a few volts. What might be the reason for this? Or is that quite normal?

Yes, that's quite normal; I've seen the same thing in my own projects. As to what causes it, it's due to fluxuations in the AC mains voltage. The bias current of the finals drifts a few mA, depending on the time of day, and the overall load on the AC distribution system.

Also, a more generic question. The volume pot becomes noisy from 50%-99% turned, becoming more and more noisy. Then at its fully on position it is dead silent. Is this poor layout or wiring? Would a stepped attenuator be the answer?

The sound still amazes me, can't believe I gave it to her! :headbash

It's dirt on the track and/or the thing is simply worn out. You might also want to take a look at the schemo to see if there is any DC flowing across the track and wiper. That'll cause a pot, especially a high resistance one, to go scratchy in no time.
 
There are "essentially" 2 versions.

From the Wiki page I've put up:
"AFAIK there are 2 major versions. The first (auto bias), which is probably the one newbies should build and a fixed bias version."

Baby Huey PP EL84 amplifier - diyAudio

I would suggest building the version in the Wiki (Which is AFAIK the latest auto bias version). And you can always try something you've lifted from this thread.
 
I guess a schematic with last tweeks (feedback scheme, zobel network, ccs, etc). I was looking to copy what has been deemed the final tweeked version. Your Wiki and other gathering of information was something I missed and will go through it. I appreciate the help. I would like to use Hammond transformers and build what is most common. Thx for the Links.....
 
Guys, sorry i've been a bit tardy on this thread of late. I hoped to give a final definitive schematic for the 2 versions. I can do that for the basic Baby Huey but not for the Version 2.

See Bas's Wiki for the basic version - this is the version I am running at home while I rebuild my Version 2 (the fixed bias with MOSFET source follower) prototype.

My basic version has (refer to the Wiki):

The tweak to add the 33 Ohms (or 39 Ohms in my unit, not critical) in the 0V side of the EL84 CCS bypass capacitors is IN. Also my bypass caps are 1000uF Blackgate PK.

The tweak to take the shunt feedback from the screen connections instead of the anode connections is OUT - I tried it for a while and then decided I liked it better the original way, I know of a few others who tried this mod and who also decided to back it out.

There is one new tweak. If you have a global feedback connection then don't try this - if you are running open loop (no global feedback) then the high frequency limit of the amp is from the diffamp stage trying to drive the Miller capacitance at the EL84 grids. You can precompensate for this limitation a bit by rolling off the shunt feedback at high frequencies. Add a Zobel (Series connected resistor plus capacitor) across the feedback set resistor. I have 16K as the shunt feedback set resistor and I have a 1K8 + 2n2 zobel across that 16K. This certainly speeds up the amp and improves its attack. Be careful with adding this zobel, there are some initially seductive settings with much stronger zobels (more capacitance) but they don't sound right after longer term listening. I found it easier to set the zobel values with a Vinyl source than by listening to CD.

As far as I'm concerned that is the final version of the basic Baby Huey. This is the "simple" version which is recommended for most constructors.

This is a lovely amp but if we want to get really picky then it has a couple of remaining warts which can be addressed at the expense of added complexity (these are the same warts which occur in 95+% of tube amps).

Wart number 1 - it has a "crap" overload characteristic. When the output tubes are overdriven into grid conduction the cathode bypass capacitors charge up. With the bias voltage being established by a Constant Current Source, recovery (bleed off that charge) is even slower than when using a standard resistor for cathode bias (with a resistor - as the voltage on the cap increases you get additional current through the bias resistor to bleed of the charge a little quicker, that doesn't happen with a CCS). The fix for this is to change to fixed bias. That can loose the simple "no adjustment required" philosophy but we could try:
1) something like the Red Light District LED bias
2) a bias servo, voltages are such that a simple op-amp circuit should cope
3) a current controlled voltage source, thats what we ideally want, I've seen a few ideas along these lines floating around here from time to time but no final schematic

Wart 2:
The high frequency limit is dependent upon the 12AX7 differential amp driving the EL84 grids Miller capacitance. to fix that we need to buffer the grid with a cathode follower or MOSFET source follower or similar. This has the added benefit of allowing the EL84 grid to be DC coupled to the source follower and having a low impedance to ground for that grid. It ties the bias point down tight and ensures fine detail is preserved by eliminating output tube grid current perturbations.
This can also be fixed by reducing the Miller Capacitance by eliminating the Ultra-linear connection and going to Pentode connection for the EL84s, unfortunately, doing that also just about guarantees that you need to introduce global feedback to get acceptable damping factor, so it is not my preferred method.

Thats where Version 2 of the Baby Huey is headed. It currently has the fixed bias and the MOSFET source follower buffer. Things I am still trying or intending to try, are:
1) Bias Servo to get back to a no adjustment required design.
2) EF86 or 6BR7 Pentodes on each side of the diff amp instead of the 12AX7 triode sections. (more gain and higher rp which will not divide the shunt feedback signal as much).

Aside: I'm also building a new amp - a quad of ElectroHarmonics 300B's driving a Plitron PAT4006 Toroidal Output Tranny for each channel (mono block) - Baby Huey'ish front end.

Cheers,
Ian
 
Last edited:
Thanks Ian,


2) EF86 or 6BR7 Pentodes on each side of the diff amp instead of the 12AX7 triode sections. (more gain and higher rp which will not divide the shunt feedback signal as much).

I have a quad of nos Brimar 6br7's (cv4006) and plenty of spare heater current available so will be looking forward to a schematic.


Brgds Bill

P.s. Trying hard not to think too much about your new amp.........:D
 
Last edited:
I am interested in building this EL84 PP design. Before I do I have some questions.

Are there some spec's available in terms of Power, THD and Zout.
The 100uF/385V Elco; I would like to change this to a MKT/MKP capacitor but it has a rather high value. Could i lower the value to 47uF or 33uF?

And does anyone know how this design compare in terms of sound to the red light destrict design?

Thanks in advance, René
 
Rene,
No specs really - I don't have the test gear to measure it.
Output Power is typically 10 Watts. That compares to 18 Watts for SY's RLD.
Zout measurements appeared in the thread some where. If I recall correctly Zout was around 1.5 Ohms. Again the RLD will better that.
I expect that the Baby Huey would have better frequency response but thats just a gut feel and I could be wrong (wouldn't be the 1st time).
Can't compare the sound since I haven't heard an RLD.

The Baby Huey has a single power rail and so the quality of that rail is important. I would be hesitant about reducing the main HV Cap too much but you could certainly try it with 47uF.

Don't forget Eli's "El Cheapo" either.

Both the "El Cheapo" and the RLD have a common cathode stage followed by a split load (concertina) phase splitter. The Baby Huey has a diffamp based splitter with balanced shunt feedback from the EL84 anodes.

"El Cheapo" usually runs triode mode output.
RLD runs Pentode Mode Output with a fair amount of global feedback.
Baby Huey runs Ultralinear Output with that balanced shunt "local" feedback and does not need any global feedback. If stereo image and sound stage are of particular importance to you then I'd recommend the BH (beware self recommendation is usually not much of a recommendation at all).

If lots of power and clean overload recovery is more important to you (for loud head banging music or dush dush music) then the RLD is probably the go. Also if your speakers are less than say 90dB/W/m sensitivity then you will need the extra power.

A comparison of the 3 circuits may give you some insights into what we were trying to achieve.

The BH design has evolved over time by pure listening. Sure I looked at the theory and what might happen if I did this or that but the determining thing for what stayed in and what was rejected was based on just listening tests not measurements (this was a deliberate policy - I could have done all the measurements, after all I'm a Electronic Design Eng. in the day job, but I went this way based upon the fact that there appears to be very little correlation between what sounds good and the sort of measurements which are "traditionally" done, i.e. THD etc).

Cheers,
Ian
 
Last edited:
Just my two pennys worth here.
I heard an implantation of the RLD and sounded very tame and uninspiring. It lasted a few months in its high end setup.
I have heard many Schade amps before and they have never disappointed> I have not heard the BH, but am more than confident that it would sound good.

I think it comes down to what is important to the designer. For SY it's the numbers and I am certain the RLD measures very well. I personally prefer to leave the numbers at home and build by ear.

Shoog
 
Thanks Ian,

I like stereo imaging an soundstage. So I think I will go for the BH.

I am a solid state "man" but I discovered tubes for the last 3 years with good results and with good satisfaction. The only thing I don't like if the sound is to smooth, to warm sounding, especially in the bass region. I can adjust the 16K cross balance resistor to my own wishes. According to this thread from SS-ish sound to a triode sound. My experience so far with tubes is (unfortunately) the better the soundstage, the softer/smoother the sound is. I am very curious how the BH will do.

Thanks again.

Best regards,
René